3D Face Analysis for Facial Expression Recognition

In this paper, we investigate the person-independent 3D facial expression recognition. A 3D shape analysis is applied on local regions of 3D face scan. The correspondent regions of different faces under different expressions, are extracted and represented by a set of closed that capture their shapes. A framework is applied to quantify the deformations between curves and compute the geodesic length (or distance) that separates them. These measures are employed as inputs to a commonly used classification techniques such as AdaBoost and Support Vector Machines (SVM). A quantitative evaluation of our novel approach is conducted on the publicly available BU-3DFE database.